> For the complete documentation index, see [llms.txt](https://docs.silk.us/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.silk.us/learn/data-management-concepts/cloud-efficiency-with-copy-data-management.md).

# Cloud Efficiency with Copy Data Management

Learn how Silk data reduction and copy data management capabilities help you support important secondary workloads while reducing cloud storage and infrastructure costs.

Organizations often need multiple copies of production data to support development, testing, reporting, analytics, AI, and other secondary workloads. In conventional cloud environments, every new copy can consume additional storage and infrastructure, increasing costs as data volumes and use cases grow.

Silk combines inline data reduction with copy data management capabilities that let you reuse production data efficiently. Teams can create data copies quickly, isolate workloads from production, and provide users with the data they need without maintaining a full physical copy for every use case.

## Reduce your data footprint

Silk DataPods continuously optimize data as it's written. These services operate across the DataPod, allowing primary data and its derivative copies to benefit from a shared, efficient storage foundation.

Silk applies the following data services:

<table data-view="cards"><thead><tr><th align="center"></th><th align="center"></th><th></th></tr></thead><tbody><tr><td align="center"><h3><i class="fa-file" style="color:$primary;">:file:</i></h3></td><td align="center"><strong>Global inline deduplication</strong></td><td>Identifies duplicate data and stores common data only once across the DataPod.</td></tr><tr><td align="center"><h3><i class="fa-compress" style="color:$primary;">:compress:</i></h3></td><td align="center"><strong>Inline compression</strong></td><td>Reduces the amount of storage consumed by compressible data in real time.</td></tr><tr><td align="center"><h3><i class="fa-eraser" style="color:$primary;">:eraser:</i></h3></td><td align="center"><strong>Zero elimination</strong></td><td>Detects empty data blocks and prevents them from consuming physical capacity.</td></tr><tr><td align="center"><h3><i class="fa-ring" style="color:$primary;">:ring:</i></h3></td><td align="center"><strong>Thin provisioning</strong></td><td>Allocates capacity as data is written instead of reserving the entire provisioned volume in advance.</td></tr></tbody></table>

{% hint style="info" %}
**Deduplication performance considerations**

Enable deduplication for each volume group as needed, based on your workload's performance requirements. Deduplication introduces a small performance impact.
{% endhint %}

Dynamic database block sizing complements these services by adapting block sizes to application workloads, helping the DataPod efficiently support different application types and concurrent workloads.

Together, these services reduce the physical capacity required to support your data. Since optimization occurs inline, applications can access data without waiting for a separate post-processing or data reduction operation.

{% hint style="success" icon="sparkles" %}
**Why it matters**

Data reduction improves more than the footprint of an individual database. Copies created from optimized data can share the same underlying resources, allowing the DataPod to support more use cases within its available cloud infrastructure.
{% endhint %}

## Create copies without duplicating all source data

Silk copy data management capabilities let you preserve, access, and repurpose data without creating a complete physical copy for every workload.

<details>

<summary><strong>Preserve point-in-time data with snapshots</strong></summary>

A snapshot captures a point-in-time representation of a volume group. Silk creates snapshots instantly with no initial data footprint, allowing you to establish recovery points or preserve data for downstream use without interrupting the source workload.

Snapshots can support:

* Recovery from accidental deletion or logical corruption
* Development and testing workflows
* Replication between DataPods
* Creation of writable data copies
* Retention of data between specific application or release states

As the source data changes, Silk preserves only the data required to maintain each point-in-time representation.

For details on consistency, scheduling, retention, and restore, see [Snapshots](/learn/data-management-concepts/snapshots.md).

</details>

<details>

<summary><strong>Provide writable data with views and thin clones</strong></summary>

Views and thin clones provide writable access to point-in-time data without immediately duplicating the source dataset. They initially share existing data and consume additional capacity only as changes are written.

This approach allows teams to create multiple isolated environments from the same data foundation. Each team can modify its copy independently without affecting the source database or another team's environment.

Use views or thin clones to:

* Provision development and test environments
* Test application upgrades and database changes
* Reproduce defects using production-like data
* Refresh reporting or analytics environments
* Experiment with new models, queries, and data pipelines
* Support parallel projects without maintaining separate full copies

</details>

<details>

<summary><strong>Clone databases with Silk Echo</strong></summary>

Silk Echo extends these capabilities with database-aware workflows. Echo coordinates operations between the database host and the underlying DataPod, allowing authorized users to create application-consistent snapshots and provision database clones through a database-oriented interface.

Echo can clone one or more databases to one or multiple hosts, providing full-performance, writable database copies in seconds. These copies can support development, testing, analytics, machine learning, and other secondary environments without placing additional demand on the production database.

</details>

{% hint style="info" icon="circle-info" %}
**Protect production-derived data**

Silk complements cloud-native encryption and supports authentication controls for management access and host-to-DataPod connections.

Apply your organization's access controls, data masking practices, and governance requirements before making production-derived data available to secondary users or systems. Silk supports the use of third-party data masking tools to help protect sensitive data.
{% endhint %}

## Choose the right copy type

Select a copy capability based on whether the workload requires point-in-time preservation, writable access, database-aware coordination, or a physically independent copy.

| Requirement                                                  | Recommended capability |
| ------------------------------------------------------------ | ---------------------- |
| Preserve a point in time                                     | Snapshot               |
| Create a writable, space-efficient copy                      | View or thin clone     |
| Clone a supported database through a database-aware workflow | Silk Echo              |
| Create a fully independent copy                              | Full clone             |
| Place data on another DataPod, region, or supported cloud    | Replication            |

## Support more workloads from the same data foundation

Space-efficient copies make it possible to reuse trusted data across teams while keeping their environments isolated.

<table><thead><tr><th width="191.111083984375">Workload</th><th>How Silk helps</th></tr></thead><tbody><tr><td><strong>Development</strong></td><td>Rapidly provision writable environments so developers can work in parallel using current, production-like data.</td></tr><tr><td><strong>Testing and quality assurance</strong></td><td>Create isolated copies for functional testing, performance validation, release testing, and defect reproduction.</td></tr><tr><td><strong>Analytics and business intelligence</strong></td><td>Move reporting and query activity away from production while maintaining access to relevant datasets.</td></tr><tr><td><strong>AI and machine learning</strong></td><td>Provide data for model development, retrieval-augmented generation (RAG), and experimentation without running those workloads against the production database.</td></tr><tr><td><strong>ETL and data processing</strong></td><td>Refresh downstream environments from a known point in time and process data without extending contention on production systems.</td></tr><tr><td><strong>Training and demonstrations</strong></td><td>Create temporary, independent environments without maintaining a permanent full copy for every user or session.</td></tr></tbody></table>

By separating these use cases from production, organizations can protect production performance while giving other teams faster access to the data they need.

## Automate recurring data workflows

Silk scheduling and automation capabilities can reduce the manual work involved in creating, refreshing, and retiring secondary environments.

Depending on the workflow, automation can help you:

* Create snapshots on a consistent schedule
* Provision development and test environments
* Refresh non-production data
* Coordinate ETL and reporting processes
* Standardize repeatable data delivery workflows
* Remove temporary copies when they are no longer required

Automation improves consistency and shortens the time between requesting data and making it available. It can also help prevent temporary environments from becoming long-running, unmanaged consumers of cloud resources.

## Create independent copies when required

Space-efficient copies are appropriate for many secondary workloads, but some requirements call for physically independent data.

Silk also supports:

{% columns %}
{% column %}

### <i class="fa-copy" style="color:$primary;">:copy:</i>

**Full clones**, which create complete, independent copies and consume capacity equivalent to the copied data.

You can create a full clone from a thin clone, and the clone is available for use immediately while Silk populates its data in the background.
{% endcolumn %}

{% column %}

### <i class="fa-arrow-left-arrow-right" style="color:$primary;">:arrow-left-arrow-right:</i>

**Replication**, which transfers snapshot-based copies between DataPods for availability, disaster recovery, workload distribution, or geographic placement.
{% endcolumn %}
{% endcolumns %}

These capabilities consume more resources than snapshots and thin clones, but they provide additional independence and resiliency. Choose the copy method based on the workload's requirements instead of creating a full copy by default.

## Balance performance, access, and efficiency

Silk allows organizations to support growing data demands without treating each new use case as a separate infrastructure deployment. Inline data reduction minimizes the footprint of stored data, while snapshots, views, thin clones, and Echo make that data available to additional workloads efficiently.

The result is a shared data foundation that can support production and secondary environments while helping teams:

* Reduce unnecessary storage consumption
* Avoid repeated full-size data copies
* Provision environments faster
* Protect production database performance
* Improve access to current, production-like data
* Scale development, analytics, and AI initiatives more efficiently

By selecting the appropriate copy type for each workload and automating its lifecycle, you can give teams access to the data they need while maintaining control over cloud resource consumption.

## Related resources

{% content-ref url="/pages/yOnS9ewZDWUWPRUHhoOT" %}
[Snapshots](/learn/data-management-concepts/snapshots.md)
{% endcontent-ref %}

{% content-ref url="/pages/dJ71QO0yxR6EjT0w1Vfw" %}
[SDP Management Console Overview](/learn/products-and-core-concepts/datapod-architecture/sdp-management-console-overview.md)
{% endcontent-ref %}

{% content-ref url="/pages/OWKl85gvHKIdZE8eGaMD" %}
[Silk Echo Overview](/learn/products-and-core-concepts/silk-product-fundamentals/silk-echo-overview.md)
{% endcontent-ref %}


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.silk.us/learn/data-management-concepts/cloud-efficiency-with-copy-data-management.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
